0
$\begingroup$

I'm running diff and diff regression using the results of a repeated cross-sectional survey.
I'm trying to test the effects of raise in min wage.
I didn't have the actual salaries of the people who answered the survey so I'm trying to use other variables in order to define my treat and control group.
I defined treat group to be people aged 17-25 who accomplished less than 12 years of education to be my treat group.
The treatment probably affected other groups in population but I defined these group as a low-skilled group with a higher chance to be affected by this treatment.

The regression I'm using is this one: $$ y_{it} = \beta_{0} + \beta_{1}\text{Treat}_{i} + \sum_{j \neq k} \lambda_{j} \text{Year}_{t=j} + \sum_{j \neq k} \delta_j \left( \text{Treat}_i \cdot \text{Year}_{t=j} \right) + X_{it}'\gamma + \epsilon_{it}. $$

Yit - is a binary variable time periods t=1,2,...,k,...,T
the treatment happens between k and k+1 (so time k is my last pre-treatment period).

Is it ok to also include "year of education variable" in X (variables which effect evenly on the treat and control group)?

$\endgroup$
7
  • $\begingroup$ Does this treatment only affect those in that age group, and with less than twelve years of education? $\endgroup$ Jun 5 '20 at 0:31
  • 1
    $\begingroup$ Please note that it's difference in difference or difference in differences. You're not adding differences. See the references here, for example: en.wikipedia.org/wiki/Difference_in_differences $\endgroup$
    – Glen_b
    Jun 5 '20 at 5:10
  • $\begingroup$ the Treatment is the raise the minimum wage. Because I do not have the payroll data of the individuals, I chose the treatment group by characteristics. Therefore, the treatment should only affect the treatment group I have chosen. $\endgroup$
    – XYZ
    Jun 5 '20 at 6:49
  • $\begingroup$ The treatment probably affected other groups in population but I defined these group as a low-skilled group with a higher chance to be affected by this treatment. $\endgroup$
    – XYZ
    Jun 5 '20 at 20:10
  • $\begingroup$ This treatment does not seem well-defined in my estimation. Is anyone with a high school education or less assumed to be unskilled? The control group consists of individuals over the age of 25 with at least some college or more. What do you hope to achieve by adjusting for education, especially when you split by education? $\endgroup$ Jun 6 '20 at 2:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.